The State of Drug Court Research

Transcription

1 research A Public/Private Partnership with the New York State Unified Court System The State of Drug Court Research Moving Beyond Do They Work?

2 Written by Amanda B. Cissner Michael Rempel 2005 This publication was supported by the Bureau of Justice Assistance, Office of Justice Programs, U.S. Department of Justice under Grant Number 98-DC-VX-K007 awarded to the Center for Court Innovation. Any opinions and interpretations expressed in this paper are those of the authors and do not necessarily represent the official position of the U.S. Department of Justice. About the Authors Amanda B. Cissner is a senior research associate at the Center for Court Innovation. Michael Rempel is director of research at the Center for Court Innovation. Acknowledgements The inspiration to write this paper grew out of several meetings with Greg Berman, Aubrey Fox, and Valerie Raine at the Center for Court Innovation, and we are grateful to all three for their initial input as well as their comments on an earlier draft. We also thank Steven Belenko, Caroline Cooper and Dana Kralstein for their helpful feedback, and Robert V. Wolf for his editorial acumen in preparing the final manuscript.

3 THE STATE OF DRUG COURT RESEARCH Moving Beyond Do They Work? Introduction When the Miami Drug Court opened its doors in 1989, it launched a dramatic shift in how courts respond to the criminal behavior of drug-addicted defendants. By combining treatment with close judicial supervision, the drug court model offers a new alternative to the unproductive and costly cycle of addiction, crime and incarceration. Unlike conventional courts, the success of drug courts is measured not by how quickly they process cases, how many convictions they produce, or how much jail time defendants receive; but on achieving tangible impacts less drug use and crime, gains in employment and education, improved mental and physical health, and cost savings from diverting offenders away from jail and prison. Their vast potential has led to a stunning national expansion over 1,300 drug courts in early 2005, less than 15 years after the Miami program enrolled its first defendant. 1 To test their performance, early drug court evaluations primarily focused on the bottom line: did they work? Most evaluations found that drug courts, while not a cure-all, produce meaningful reductions in re-offending compared with conventional prosecution. 2 The combination of favorable results and massive, ongoing efforts to open new drug courts nationwide has now spawned an urgent set of second-generation questions focusing less on whether drug courts work and more on how and for whom, along with how they might work better. These are action research questions. Action research focuses less on evaluating bottom-line success and more on providing feedback that can improve everyday program quality. Among the topics that drug court action researchers are currently investigating are: Target population Which categories of participants (e.g., based on drug use and treatment history, criminal history, charges, socioeconomic variables, mental health, or other factors) are especially likely to benefit from drug court? Are today s programs reaching and enrolling the ideal target population? Program components How important is each component of the drug court model (e.g., team approach, treatment, case management, judicial supervision, rewards, and sanctions)? How is each component best administered? Quality of treatment Which treatment modalities are most appropriate for different categories of participants, and are such modalities widely available? How can 1

4 Center for Court Innovation drug courts better monitor the quality of the treatment services on which they depend? Drug court retention and graduation How long should drug court participants be retained in the program in order to benefit from the intervention? What steps can be taken to improve retention? How important is drug court retention and graduation in achieving positive long-term outcomes? The answers to these questions can be used by drug court practitioners to refine their practices and apply resources more wisely. States engaged in largescale institutionalization efforts can incorporate known best practices into statewide drug court protocols. Indeed, action research may be critical to the longterm sustainability of drug courts. Without understanding which of the key drug court components have the greatest impact, and which categories of participants will benefit most, newer drug courts coping with fewer resources and lacking the charismatic leadership of early pioneering judges may not know which parts of the model must be preserved intact and which can be tinkered with. This may lead the success of drug courts to slip hand-in-hand with institutionalization. Although the early 2000s have seen the completion of many valuable studies, some of the most interesting results have yet to be widely disseminated. This paper synthesizes some of the more revealing national findings and highlights areas where we need to know more. Although findings discussed here typically required extensive data collection and evaluation expertise, drug courts can also do a great deal on their own, with modest investments in data collection and analysis. A companion paper discusses how local drug courts can start their own action research program, using simple and easy-to-collect data to answer practical questions about their volume, participant characteristics, and performance. 3 This paper provides a broader context by focusing on general lessons learned. The Drug Court Model: What Are We Evaluating? What do drug court evaluations typically report? In most examples to date, the evaluation found that drug court participants (including both graduates and failures) had lower recidivism rates than similar defendants prosecuted with conventional methods. Some evaluations have considered other outcomes, such as drug use, employment, health care, time spent in jail and prison, or cost savings; but most have examined effects on recidivism alone mainly because the availability of official criminal justice records makes recidivism analyses the easiest and least costly to implement. As important as it is to know that drug courts reduce recidivism, this is not nearly enough. The figure on the next page illustrates how little we learn merely from knowing that a drug court, in its totality, reduced re-offending. More helpful at this point would be research telling us which specific drug court components made the greatest difference in producing successful outcomes, and which, if any, made no difference. Perhaps a drug court receiving a positive evaluation would have been equally effective with a model excluding rewards and sanctions; or excluding case management; or 2

5 The State of Drug Court Research excluding treatment while relying only on court-based judicial supervision. Or perhaps the drug court would have produced even better results if certain policies were administered differently, for example if in-court interactions with the judge were longer or more probing; or if more effective treatment modalities were used; or if drug testing was more frequent. Consider just a few of the specific components that are believed to underlie drug court success (see the third column in the figure): Early Identification Drug court participants are believed to be most receptive to change at the crisis moment of the arrest (or outset of the case); hence potential participants should be identified, assessed, and placed in treatment as rapidly as possible. Community-Based Treatment Effective substance abuse treatment modalities are believed to exist that can promote sobriety; hence drug courts should match participants to appropriate community-based residential or outpatient treatment programs. However, the science of treatment-matching is not very well-developed; there is little evidence indicating which treatment interventions work best and with which populations. Legal Coercion Drug court participants are believed to be retained in treatment at higher rates than those entering voluntarily because drug courts offer concrete legal incentives to do well namely, the prospects of a charge reduction or case dismissal in the event of graduating and jail or prison in the event of failing. THE DRUG COURT MODEL Background Context Participants The Drug Court Outcomes Local Environment Demographics Drug Use Patterns Political/Legal Culture Stakeholder Attitudes (D.A., defense bar, judges, and treatment) Drug Court Eligibility (charges, criminal history, and addiction severity) Participant Profile Demographics Socioeconimic status Community/family ties Drug use/treatment history Criminal background Drug Court Components Early Identification Treatment Legal Coercion (e.g., threat of jail) Judicial Interaction Rewards Sanctions Team Approach Case Management Drug Testing Supplemental Services (employment, mental health, housing, etc.) Community Outreach Information/Evaluation Recovery Motivation and Behavioral Change Drug Court Graduation Rehabilitative Outcomes Reduced Crime and Drug Use Other Outcomes (e.g., employmentand health-related) Social Benefits Public Safety Less Incarceration Cost Savings 3

6 Center for Court Innovation Judicial Supervision Through regular status hearings before the drug court judge, in which the judge engages in direct conversation with the participant about progress and setbacks, the judge is thought to play an instrumental role in promoting sobriety. Rewards and Sanctions Drug courts are believed to encourage progress by applying a continuum of intermediate sanctions and rewards; sanctions are thought to be most effective when applying principles of certainty (each infraction receives a sanction), celerity (sanctions are imposed as soon as possible after the infraction occurs), and severity (sanctions rise in severity in response to repeat infractions and consider the severity of the behavior). Team Approach Drug courts are believed to be more effective when the parties (e.g., judge, lawyers, and clinical staff) curtail the adversarial process and work together to figure out what will best promote the recovery of each participant. Official drug court publications and everyday practitioners hold all of these components to be important, but are any backed by evidence? This paper considers what we know now concerning: (1) whether drug courts work, (2) how the work, and (3) for whom. The first question, while the most researched, is covered only briefly, given this paper s focus on probing deeper to what lies behind the more widely known findings. Nearly all results presented here pertain exclusively to adult drug courts, since research is only beginning to emerge on the family and juvenile drug court models. Do Drug Courts Work? Before tackling the more challenging questions of how and for whom drug courts work, what does the literature show concerning their overall success? (See fourth column of the figure.) It is worth noting here that much of the literature to date has been plagued by methodological issues necessitating the careful interpretation of many drug court studies. For instance, in a review of the literature, Steven Belenko points to a lack of precision in defining data sources, timeframes, and measures as well as data quality issues and missing data in many studies. 4 Additionally, several studies have relied on inappropriate comparison groups (e.g., drug court graduates compared to drug court failures; drug court participants compared to those who were found ineligible for the program) or have used no comparison group. 1. Treatment Retention Drug court retention rates far exceed those for the general treatment population. Retention is a key measure of program success. A one-year retention rate, for example, indicates the percentage of participants who, exactly one year after entering drug court, had either graduated or remained active in the drug court program. Earlier research finds that longer retention not only indicates success in treatment but also predicts continued future success in the form of lower post-treatment drug use and criminal offending. 5 4

7 The State of Drug Court Research Drug courts have been consistently found to produce higher retention rates than community-based treatment programs accepting a combination of voluntary and court-mandated treatment participants. 6 This is believed to be due in part to the legal pressure entailed by the threat of incarceration drug court participants face in the event of failure; several studies confirm that legal coercion is a sizable force improving both short-term and long-term treatment outcomes. 7 As for hard numbers, one review estimates that drug courts nationwide have an average one-year retention rate of 60 percent. 8 A study of 11 New York State drug courts found a slightly higher median one-year retention rate of 66 percent; and estimated graduation rates exceeded 50 percent in eight of 11 sites. 9 On the other hand, a study of four mentor drug courts in other states reported an average graduation rate of only about one-third. 10 Nevertheless, every one of these estimates improves considerably upon those obtained at community-based treatment programs, where many participants enter voluntarily without the pressure of a court mandate. Nationwide, approximately half of those enrolling in outpatient treatment are retained for less than three months. 11 Since attrition always increases over time, oneyear retention rates across these same programs would presumably drop much lower. Indeed, focusing on therapeutic communities (involving residential treatment), one study reports one-year rates ranging from just percent, lower than even the very worst performing drug courts. 12 So drug courts have clearly achieved success in keeping addicted persons in treatment for longer than other treatment models. 2. Recidivism Adult drug courts significantly reduce recidivism, although the level of impact varies over time and by court. In their comprehensive review, David Wilson and colleagues reported that 37 of 42 studies found lower recidivism rates among drug court participants than comparison groups composed of similar but non-participating defendants. Most of the studies defined recidivism as re-arrests, some as re-convictions. The average effect size was approximately 13 percentage points. 13 Other literature reviews that considered fewer total drug court evaluations, mainly by eliminating ones with particularly weak methodologies, still reported lower recidivism rates among drug court participants than comparison group defendants in nearly all sites. 14 An important caveat to these results is that most studies only examined recidivism over a brief time frame, usually coinciding with the in-program period of active drug court participation. Only a handful extended the measurement period beyond two years after program intake. A study of the Baltimore City Treatment Court, which used a strong research design where defendants were randomly assigned either to the drug court or conventional case processing, tracked defendants over a third year and found sustained differences, with a recidivism rate that was 10 percent lower for drug court participants than the comparison group. 15 Several studies have been able to isolate recidivism over a post-program period after drug court participation has ended (i.e., after the date of graduation or failure/release from incarceration). A study of six New York State drug courts, for example, reported consistent recidivism reduc- 5

8 Center for Court Innovation tions over a one-year period after graduation or failure an average 31 percent reduction in relation to the comparison group during a comparable one-year post-disposition period. When focusing on graduates alone, the impact is truly staggering an average 71 percent reduction across the same six sites. On the other hand, drug court failures were as or more likely as comparison group defendants to re-offend. This means the benefits of the drug court accrue primarily to those who successfully complete; therefore, to have a substantial net impact when averaged across all participants, graduating a significant percentage may be extremely important. 16 Overall, qualifications concerning methodology and measurement periods notwithstanding, results to date offer strong support for drug courts. Indeed, the U.S. Government Accountability Office (GAO) concluded definitively in its 2005 report that adult drug courts succeed in reducing recidivism Drug Use Studies show varying levels of continued drug use among drug court participants. Many comparison groups are not tested for drug use. Few studies directly measure reductions in drug use, primarily due to the inherent difficulties in locating both drug court participants and comparison group members for follow-up interviews and urinalysis testing. In one study, participants in Maricopa County, Arizona were found less likely than defendants randomly assigned to a regular probation group to test positive for heroin or cocaine one-year after program entry; however participants were found more likely to test positive for marijuana. 18 Two other studies found that drug court participants were significantly less likely than comparison group defendants to use several illegal substances after a short follow-up period. 19 The GAO aptly concluded that results are limited and mixed when it comes to effects on drug use. 4. Other Rehabilitative Outcomes Virtually no studies measure other outcomes such as employment, welfare dependence, and mental or physical health. Some evidence suggests that drug courts may produce modest gains in these areas. Studies of the Baltimore City and Brooklyn drug courts detected few significant social and economic impacts, such as reductions in family, psychiatric, medical, or employment problems. However, where differences were evident, the tendency was for drug court participants to show modest relative improvement on these types of measures. Clearly, more research is needed here. 5. Cost Savings While few studies measure cost impacts, nearly all of the available evidence demonstrates that drug courts save money over the long-term. Most studies considering cost savings have focused on savings to the criminal justice system (e.g., courts, corrections, probation, or prosecutors). These savings are the easiest to quantify but not necessarily the largest, as compared with others, such as reduced taxpayer-funded health care costs and emergency room visits, reduced dependence on public assistance, and savings to the community through reduced victimization costs. According to two recent reviews of the literature, nearly all complet- 6

9 The State of Drug Court Research ed cost studies show significant net savings. 20 Of the completed studies, the most noteworthy are two statewide evaluations of drug courts in Washington and California. The Washington State study found savings of $3,892 per drug court participant; or savings of about $1.74 for every dollar invested. 21 The California study reported average yearly savings of $2,000 per participant, though results varied widely across six separate drug court sites. Two California sites produced per participant savings in excess of $15,000 while, on the other end of the spectrum, one produced net costs of just over $9,000 (this was the only site that failed to save money on net). 22 Since many of the savings stem from reductions in recidivism (the savings arise because justice system agencies do not have to deal with future cases), drug courts that achieve larger reductions in recidivism will naturally produce larger cost savings. Further, since recidivism-related savings accrue in the long-term, one should acknowledge that the immediate up-front costs to the court system of running a drug court generally exceed those of conventional case processing. Other justice system agencies, such as the District Attorney s Office, the defense bar, or corrections may see more immediate cost efficiencies, particularly in those drug courts managing to reduce incarceration time on the initial drug court case. Still, it is clear that drug courts do not always produce a short-term budgetary payoff and should rather be viewed as an investment in the future. 6. Reduced Use of Incarceration As an alternative to incarceration, drug courts typically aspire to reduce the time that defendants spend in jail or prison. Limited data indicates that this happens to some degree, but not always. Some drug court critics argue that, due to the lengthy jail or prison sentences commonly imposed on drug court failures, when considering all drug court participants together, they face more severe criminal justice sanctions on average than conventional prosecution. 23 Indeed, the study of the Baltimore drug court found that while participants spent fewer days than the comparison group in jail due to their sentence, they spent substantially more time in jail due to intermediate sanctions for noncompliant behavior. Therefore, when all time was considered, the total number of days that drug court participants spent incarcerated was only slightly lower than for the comparison group. 24 In the New York State study, drug court participants in three of six sites averaged significantly fewer days incarcerated than the comparison group on the initial case; but participants in one court spent significantly more time incarcerated and in the final two sites, there was not a significant difference in either direction. 25 (Of course, since drug courts reduce recidivism, it is likely that if including incarceration time served as a result of new offenses, most drug courts would ultimately achieve reductions in net jail or prison time.) Further breaking down the results in the New York study, it bears emphasizing that drug court graduates were never incarcerated as part of their final sentence; therefore, graduates gained the full benefit of the alternative to incarceration opportunity. On the other hand, in four of six New York sites, drug court failures averaged significantly longer sentences than the comparison group. This again underscores 7

10 Center for Court Innovation the critical role of drug court graduation determining whether or not participants will benefit from the intervention. How Do Drug Courts Work? Here the evidence is more limited. The third column of the figure included above identifies 12 drug court components, of which notable evidence exists bearing on seven. 1. Early Identification Those drug court participants who are identified and begin treatment quickly are more successful than those whose entry into a community-based treatment program is delayed. A growing body of research suggests that immediate engagement is critical. Participants engaged early in the drug court process, often measured by whether a participant actually begins attending treatment within the first 30 days after formally agreeing to enter a drug court, are more likely to be retained and to have successful long-term outcomes. 26 Implication: Drug courts should strive to implement formal, streamlined intake procedures that can move potential participants rapidly from screening and assessment to formalization of participant status to placement in a suitable community-based treatment program. Where treatment slots are difficult to locate, or systematic delays in case processing cannot be overcome, this may hinder a drug court s effectiveness. To compensate, strategies such as holding pre-placement groups onsite at the drug court may help to keep participants engaged while they wait for a community-based treatment slot to become available. 2. Treatment Some contend that treatment per se does not contribute to the overall effectiveness of drug courts and that, instead, judicial supervision makes the greatest difference. Contrary to this position, evidence indicates that treatment can make a difference; but little is known about the relative impact of different treatment modalities; or about which modalities are most appropriate for different categories of participants. In the drug court world, while it may be sacrilege to label treatment as irrelevant, Mark Kleiman believes that the limited scope and duration of the drug court coupled with high costs ultimately restrict the potential impact of this intervention. Therefore, in the interest of achieving the most comprehensive impacts, Kleiman argues that the drug court model could be replaced with a bare-bones approach requiring substance abstinence reinforced through drug screening and guaranteed sanctions for noncompliant behavior but excluding a requirement of attendance in communitybased treatment. 27 Others respond that treatment itself is essential. Many studies (though not specifically of drug court participants) confirm that more time in treatment leads to more positive post-treatment outcomes on measures such as drug use, criminal activity, and employment. 28 The Baltimore drug court study confirms that participants who completed more total days in treatment reported less illegal drug use than others three years after program entry. 29 While the literature confirms that treatment is important, it is unclear whether drug courts use the most effective treatment modalities and programs. What is 8

11 The State of Drug Court Research known is that practices vary widely. For example, across 11 New York State drug courts, the percentage of participants initially referred to a residential treatment program ranged from 1 percent to 53 percent. 30 In part, this variation reflects the desire of some drug courts to keep participants in the community, rather than sending them to a residential facility. Not only do practices between courts vary; information concerning best practices is limited. For example: What modality is most appropriate for different categories of participants (e.g., severely addicted heroin users, young marijuana users, or addicted women with children)? How much treatment is ideal and do drug courts that require treatment stays well in excess of one year encounter a point of diminishing returns? (In recent years, several drug court researchers, including one of this article s coauthors, have suggested that it may be counterproductive to keep participants enrolled for too long before allowing them to graduate.) Within each basic type of modality (e.g., residential, short-term rehabilitation, or outpatient), are quality treatment services available? Are treatment providers using methods found to be therapeutically effective? How can drug court staff assess the quality of available treatment? Do variations in treatment program quality tangibly affect participant outcomes? From a comprehensive review of the treatment literature, Faye Taxman synthesizes existing treatment knowledge and recommends certain specific practices (e.g., cognitive behavioral approaches, matching defendants to appropriate programs, and clinical assessments) as crucial to successful outcomes. 31 At the same time, Taxman laments that treatment programs serving drug court participants tend to spend relatively little time less than 20 percent in one study addressing clinical issues with an approach known to be effective. 32 Several other researchers participating in a recent roundtable discussion echoed concerns that programs available to drug courts did not generally use the most effective of available modalities. 33 And drug court participants themselves criticized the quality of their treatment in several recent focus groups. 34 While it is debatable how much control drug courts can actually exert over treatment administered in community-based programs, these findings at least raise concerns about whether drug court effectiveness might be greater if the average quality of treatment was improved. 3. Legal Coercion Legal coercion can increase the incentive for drug court participants to succeed. As discussed previously, part of the success of drug courts in retaining partici- 9

12 Center for Court Innovation pants is believed to stem from the legal coercion entailed by the threat of incarceration for failing. Further, some evidence indicates that added amounts of legal coercion within drug courts can produce incrementally better outcomes. For instance, presumably because of the added leverage that results when participants are required to plead guilty in advance of participation, drug courts using post-plea as opposed to pre-plea models may be more effective. One study of a court-mandated treatment program (not a drug court per se) confirmed that the program s one-year retention rate rose by 10 percent (64 percent to 74 percent) after switching from a pre-plea to post-plea model. 35 Also, comparing different post-plea situations, a study of the Brooklyn Treatment Court found that participants facing a progressively longer jail or prison sentence in the event of failing were increasingly likely to become engaged in treatment. 36 However, the recent statewide study in New York found less strongly supportive evidence for this relationship across a range of drug courts, and the Baltimore study did not confirm this relationship at all. 37 A study of the Las Vegas drug court similarly found that clients entering the drug court post-plea performed worse than pre-plea clients, but the authors believed this may be due to a higher risk clientele entering the court post-plea. 38 Thus while the coercive aspects common to all drug courts are effective when compared with voluntary treatment, further research is needed to clarify under what conditions extra levels of coercion produce added value in terms of additional improvements in participant outcomes. In this regard, the work of Doug Young and Steven Belenko is highly suggestive. In one study, they found that treatment retention rates varied as a direct result of variation on four distinct legal coercion dimensions: (1) information (degree to which program rules and consequences of noncompliance were clearly communicated to participants); (2) monitoring (degree to which compliance was closely monitored through regular progress reports to the court and other means); (3) enforcement (degree to which noncompliant participants could expect to be rapidly caught, brought back to court, and face consistent consequences); and (4) severity (length of the resulting jail or prison sentence or other consequence). Further, this research suggests that legal coercion becomes more effective when coupled with clear communications by justice system authorities that reinforce participants impression that failure will elicit adverse consequences. This reinforcement creates a perception of coercion, which in turn mediates the relationship between the court s objective mandate on one hand and the resulting compliance outcomes on the other. For example, the dimension of severity is not measured merely by the objective facts of what will happen if participants fail but by participant perceptions of how much jail or prison time or what other consequence they will face. Implication: Drug courts should convey clearly, frequently, and specifically to participants exactly what will happen if they graduate (case dismissal or other legal benefit) and what will happen if they fail (how much jail time they will have to serve); and should convey the nature of the court s monitoring and enforcement efforts to detect and address noncompliance

13 The State of Drug Court Research 4. Judicial Supervision Ongoing judicial supervision by the drug court judge works with high-risk drug court participants. Research suggests that judicial status hearings especially ones that include positive feedback from the judge and that focus on high-risk participants can be effective. A series of random assignment studies found that drug court participants diagnosed as having antisocial personality disorder and/or having previously failed a drug treatment program did significantly better when required to appear biweekly before the drug court judge. On the other hand, low-risk participants who did not have these characteristics did either similarly or worse across different drug court sites when monitored biweekly. 40 Implication: Scarce judicial supervision resources are best targeted to high risk participants. Since participants who do attend status hearings often develop a relationship with the judge, some research has found that it can be damaging when one judge replaces another. A study of the Portland drug court reported declining treatment attendance after it switched from a single, dedicated judge to a judicial rotation system involving frequent changes in the presiding judge. Likewise, the study found that participants appearing before a single judge were less likely to be terminated unfavorably from the drug court program than participants appearing before multiple judges in the course of their participation. 41 Concerning the content of effective status hearings, a Broward County, Florida study found that in general, supportive comments by judges resulted in fewer subsequent positive drug screens, while adverse comments had the opposite effect. 42 And further confirming the importance of positive feedback, interviews with participants in two different drug courts using the same scales both found that participants rated praise from the judge and direct interaction with the judge as among the most useful drug court components. 43 Similarly, participants offering feedback in two separate focus group studies spanning nine drug court sites consistently underlined the motivating role of praise and approval from the judge. 44 By contrast, the overriding prevalence of negative and stigmatizing judicial feedback was held largely responsible for the negative evaluation results (higher rates of re-offending among participants than the comparison group) in one study of the Las Vegas drug court Rewards Rewards appear effective when they are tangible and applied frequently throughout the drug court participation process; but the literature is limited. As noted above, several studies cite the importance of positive judicial feedback. Whether more tangible rewards such as tokens, journals, or gift certificates matter is, however, a different question. Classic behavioral modification techniques of course recommend the liberal use of rewards. Yet, only one study tests the impact of rewards in drug courts with a rigorous research design. For this reason, caution is still advised before making strong assumptions about the degree to which rewards make a difference. In the one completed study, Doug Marlowe and his colleagues randomly assigned participants in one drug court to one of three rewards schedules: 11

14 Center for Court Innovation Standard rewards: hat or candle after three months of compliance; reduced community service after six months; reduced drug testing after seven; reduced treatment requirements after eight; reduced judicial status hearings after 10; and reduced homework assignments after 11. Enhanced graduated rewards: gift certificates after each additional month of compliance (i.e., 12 total certificates) that begin at $5 after month one and grow to $60 after month 12. Enhanced thinning rewards: gift certificates of $30 after months one, two, and three of compliance; $50 after month five; $75 after month nine; and $125 after month 12. Within one year of the random assignment, the graduation rate and Phase Four completion rate (the drug court used four phases) were significantly higher for participants on the two enhanced schedules than for participants on the standard schedule. In particular, 55 percent and 60 percent respectively of participants on the two enhanced scheduled had at least completed Phase Four by the one-year mark, but only 26 percent of those on the standard schedule had done so. 46 Implication: Based on this study, rewards are effective, but not in the way they are traditionally implemented in drug courts; instead, participants respond better when the rewards have tangible value, are administered more frequently throughout participation, and are administered in escalating quantities. 6. Sanctions Drug court sanctions appear effective when applied consistently and fairly; but the literature is limited. Following classic behavior modification principles, sanctions and rewards have always been core components of the drug court model. In one study of the Washington, D.C. Judicial Sanctions Program, defendants assigned to receive sanctions in response to noncompliance were less likely to be rearrested than a second group of defendants assigned to receive regular drug testing but without judicial monitoring or sanctions. 47 But little is known about precisely why or how sanctions work which specific types of sanctions are most effective, under what circumstances, and how much of a difference they truly make in a drug court setting. It is possible that intermediate sanctions are less important in drug courts than other aspects of judicial supervision, such as probing and positive judicial interactions with participants and the overarching incentive created by the threat of jail or prison for failing (see above). And because some research has shown that negative, stigmatizing in-court interactions can adversely affect subsequent performance, intermediate sanctions must be administered with care. Where similar sanctions are consistently applied in response to similar behaviors, and where the judge clearly articulates the reasons for imposing each sanction, participants may become more likely to respond positively. In general, research shows that where defendants believe justice system authorities have treated them fairly and with respect, they are more likely to comply 12

15 The State of Drug Court Research with court orders. 48 Implication: Drug courts need to cultivate a sense that their sanctioning process is fair. Developing and consistently implementing a formal graduated sanctions schedule may be helpful in this regard. 7. Team Approach The impact of the team approach has not been rigorously tested, but drug courts appear to function better when a non-adversarial team model is present. Most drug courts hold regular case conferencing meetings including the judge, attorneys, treatment providers, and other affiliated staff. In these meetings, disparate goals (e.g., of opposing attorneys) are supposed to be put aside to promote the recovery of each participant. While it is difficult to quantify the impact of a team model, a couple studies suggest it may be important. In one involving focus groups with judges from drug courts and other problem-solving courts (e.g., domestic violence courts, mental health courts, community courts), judges repeatedly cited the team approach as among the most critical ingredients for their programs to be effective. 49 Also, a recent process evaluation of the Staten Island, New York drug court found that the strong personal and working relationships established among team members the judge, prosecutors, and assigned defense counsel especially enabled the court to successfully address multiple implementation challenges during the planning stages and first year and a half of operations. 50 Still, these studies by no means involved rigorous, carefully designed tests of the team model. 8. Other Drug Court Components There is little or no evidence on the role of case management, drug testing, community outreach, and supplemental services in areas such as employment, housing, or mental health. Concerning case management in particular, today s drug courts exhibit considerable diversity of practice. Different drug courts range from employing their own onsite case management team (the most costly option); to collaborating with local departments of probation to perform case management; to eliminating case management services altogether and folding their functions under substance abuse counselors at assigned treatment programs. 51 Yet it is entirely unclear which, if any, of these approaches is more or less effective than any other. 9. Graduation Participants who reach graduation are more likely to attain continued success thereafter. What role does drug court graduation play in producing long-term rehabilitative outcomes (see middle of fourth column in the figure above)? Can those who fail drug court nonetheless gain from the experience? Several studies suggest they cannot that graduation is a pivotal milestone and that without it continued progress is unlikely. 52 For example, the New York study found that across six drug courts, there was consistently no additional benefit gained from completing more time in the drug court program only to fail in the end. Among those who failed, more time enrolled in the drug court or attending treatment prior to failure had no impact on future re- 13

16 Center for Court Innovation offending rates. Implication: Graduation is the key to successful long-term outcomes; drug courts should seek to graduate a meaningful percentage of their participants. For Whom Do Drug Courts Work? Are some components or the drug court model overall more effective with some categories of participants than others? If policy-makers knew who benefits the most, they could adjust their target populations appropriately. By contrast, consider how a drug court s target population is typically defined now (see first column of the figure above). Local community values and the specific attitudes of powerful stakeholders (e.g., the prosecutor, defense bar, and judiciary) inevitably influence a drug court s eligibility criteria: e.g., whether it will admit defendants charged with felonies or misdemeanors only; whether drug sales charges are eligible; and what type of prior criminal history is acceptable. The drug court population also reflects the character of the local population (e.g., the racial makeup or income distribution; and the nature of the local drug problem). In addition to these factors, in the ideal, the target population would be defined at least in part based on hard evidence concerning which categories of participants (e.g., based on drug use, treatment, or criminal history, sex, age, race, socioeconomic background, or other factors) are most likely to benefit from the intervention. 1. Categories of Defendants Most Likely to Perform Better in Drug Court than in Conventional Court Little is known about which categories of defendants are most likely to benefit from the drug court intervention; but three have emerged as likely candidates: (a) high risk offenders, (b) those facing greater legal consequences for failing drug court, and (c) drug offenders (as opposed to offenders arrested for property crimes and other offenses). A number of studies indicate that high risk offenders are especially likely to benefit from the drug court model. As noted above, biweekly judicial status hearings had a positive impact on drug court participants with previous failed treatment episodes and/or anti-social personality disorders; while biweekly status hearings made no difference or even a negative difference for other categories of participants. Also, a study of the Los Angeles drug court found that while the drug court did not produce significantly different re-arrest rates from the comparison group among low risk defendants, it did generate considerably lower re-arrest rates among both medium and high risk defendants. In this study, high risk defendants had more serious criminal records and weaker community ties. 53 Complementing this research is the finding cited above that participants perform better if their offenses were more serious and hence face more severe legal consequences if they fail. When comparing those processed through the drug court with those processed through conventional methods, it turns out that the drug court makes a greater relative difference in reducing the likelihood of re-offending for those with a prior criminal record. Implication: drug courts produce better outcomes if they expand their eligibility criteria to defendants with a prior criminal record, previous failed treatment, and other risk factors; conversely, limiting the drug court opportunity to less serious types of offenders (as many jurisdictions have chosen to 14

17 The State of Drug Court Research do) will reduce program efficacy. In particular, courts accepting participants over whom they can exercise more legal coercion stand to produce better outcomes. Finally, a study of one drug court found that it produced a relatively greater reduction in recidivism for defendants entering on drug than on non-drug charges, such as property offenses or prostitution. 54 In general, drug courts may work better at reducing crime related to drug use and addiction but relatively less well with crime driven by other criminal impulses or motivations. And while many property offenders may be seeking to support an addiction, it is possible that on average, crimes committed by property offenders are less often driven by an addiction and more often by other criminal propensities. Of course, this relationship should be interpreted with caution until additional research replicates the finding. 2. Categories of Defendants Likely to Perform Well in Either Drug Court or Conventional Court Certain categories of defendants are likely to perform well both in and outside the drug court: those who (a) are older, (b) have no prior criminal record, (c) abuse a primary drug other than heroin and cocaine, (d) have no dual diagnosis, and (e) have higher socioeconomic status. There are certain categories of defendants who are likely to perform well or poorly whether they are in drug court or not. Those who have a tendency to do poorly are not necessarily inappropriate for drug court; however, they may be candidates for extra monitoring or services. Synthesizing a large number of studies emerging over the past five years, some of the personal characteristics found to increase the probability of success (whether in a drug court or not) include: Older age; 55 No prior criminal record; 56 Primary drug is not a hard drug e.g., heroin and cocaine; 57 No dual diagnosis is present (major depression, bipolar disorder, or suicidal ideation); 58 and Higher socioeconomic status (e.g., as measured by educational attainment or employment status). 59 Therefore, defendants lacking the above characteristics are more likely than others to require extra attention, although more research is needed to determine exactly what kinds of interventions are most effective. 59 Conclusion The past several years have seen a remarkable convergence of support throughout the research community around the effectiveness of drug courts. Recent opinion pieces by John Goldkamp, Adele Harrell, and Doug Marlowe, three of the most prominent researchers in the field, all conclude that adult drug courts have been proven effective, a conclusion further echoed in an early 2005 report by the Government Accountability Office. Marlowe adds for emphasis, We know that drug courts outperform virtually all other strategies that have been attempted for drug-involved offenders. 60 Alongside this bold endorsement, however, Marlowe 15

18 Center for Court Innovation offers an equally important caution, Some components [of the model] may be indispensable, others may not be worth the cost, and still others may have negative side effects. 61 Indeed, the future of drug courts may well depend not on producing additional studies demonstrating their effectiveness overall but on increasing our understanding of which components are critical, which are not, and for which categories of participants the intervention works best. Information concerning the essential ingredients of drug courts becomes all the more important in light of recent efforts by many states to go to scale, expanding the reach of drug courts to far greater numbers of defendants. These efforts appear to fall under two possible paths. One involves increasing the number of drug courts as well as attempting to boost the volume of defendants served by each one. Many states, including California, Florida, Louisiana, Missouri, and New York, have already begun proceeding in this manner through statewide coordination and expansion efforts. 62 The second possible path involves applying problem-solving practices (i.e., the set of practices common to drug and other specialized courts) outside the specialized court setting throughout conventional courts. 63 While this second path has not been undertaken systematically by any state, interest in this area is growing. Both paths of expansion stem in no small part from the documented success of adult drug courts. Yet, it is unclear whether broader institutionalization can produce equally positive results. It is possible that a drug court approach will not work as well if institutionalized throughout greater numbers of courtrooms, targeting a wider range of defendants, and requiring the collaboration of far more judges, attorneys, and other stakeholders who may not all possess the dedication or skill of the movement s pioneers. This paper has sought to inform the future of the drug court movement by summarizing what we know so far about successful drug court implementation. While the research cannot give policymakers a blueprint for how to go to scale with drug court, it does offer a number of helpful lessons that will hopefully spark new thinking about how to ensure that courts are making a difference in the lives of addicted offenders. 16

19 The State of Drug Court Research Notes 1. Office of Justice Programs, BJA Drug Court Clearinghouse and Technical Assistance Project, American University, 1,367 drug courts open as of March 29, See D. B. Wilson, O. Mitchell, and D. L. MacKenzie, A Systematic Review of Drug Court Effects on Recidivism, draft M.S. (2003). See also the conclusions of Government Accountability Office, Adult Drug Courts: Evidence Indicates Recidivism Reductions and Mixed Results for Other Outcomes, United States Government Accountability Office, Report to Congressional Committees (February 2005); J. Goldkamp, The Impact of Drug Courts, Criminology and Public Policy 2: 2: (2003); A. Harrell, Judging Drug Courts: Balancing the Evidence, Criminology and Public Policy 2: 2: (2003). 3. See M. Rempel, Action Research: Using Information to Improve Your Drug Court, New York: Center for Court Innovation (2005). 4. S. Belenko, Research on Drug Courts: A Critical Review 2001 Update, National Center on Addiction and Substance Abuse, Columbia University (2001). 5. M. D. Anglin, M.-L. Brecht, and E. Maddahian, Pre-treatment Characteristics and Treatment Performance of Legally Coerced Versus Voluntary Methadone Maintenance Admissions, 27 Criminology, 537, 556 (1989); G. DeLeon, Legal Pressure in Therapeutic Communities, 18 Journal of Drug Issues, 625, 640 (1988); R. Peters and M. Murrin, Evaluation of Treatment-Based Drug Courts in Florida s First Judicial Circuit, Tampa, FL: Department of Mental Health Law and Policy, Louis de la Parte Florida Mental Health Institute, University of South Florida (1998); F. Taxman, Reducing Recidivism Through A Seamless System of Care: Components of Effective Treatment, Supervision, and Transitional Services in the Community, Greenbelt, MD: Washington/Baltimore HIDTA Project (1998); F. Taxman, B. Kubu, and C. DeStefano, Treatment as Crime Control: Impact of Substance Abuse Treatment on the Individual Offending Rates of Hard-Core Substance Abusing Offenders, Greenbelt, MD: Washington/Baltimore HIDTA Project (1999). 6. W.S. Condelli and G. DeLeon, Fixed and Dynamic Predictors of Client Retention in Therapeutic Communities, 10 Journal of Substance Abuse Treatment (1993); B.F. Lewis and R. Ross, Retention in Therapeutic Communities: Challenges for the Nineties, Therapeutic Community: Advances in Research and Application, eds. F. M. Tims, G. DeLeon, and N. Jainchill. NIDA, Rockville, MD (1994). 7. M. D. Anglin, et al., supra note 5; G. DeLeon, supra note 5; M. Hiller, K. Knight, and D. D. Simpson, Legal Pressure and Treatment Retention in a National Sample of Long-Term Residential Programs, Criminal Justice and Behavior 25: (1998); D. Young and S. Belenko, S., Program Retention and Perceived Coercion in Three Models of Mandatory Drug Treatment, Journal of Drug Issues 22: 2: (2002). 8. S. Belenko, Research on Drug Courts: A Critical Review, 1 National Drug Court Institute Review 1 (1998). 9. M. Rempel, D. Fox-Kralstein, A. Cissner, R. Cohen, M. Labriola, D. Farole, A. Bader, and M. Magnani, The New York State Adult Drug Court Evaluation: Policies, Participants, and Impact. Report submitted to the New York State Unified Court System and the Bureau of Justice Assistance, New York: Center for Court Innovation (2003). 10. D. F. Anspach, V. Collom, and A. Ferguson, Assessing the Impact of Functional Components of the Drug Court Model on Post-Program Recidivism. Presented at the Annual Conference of the American Society of Criminology, Denver, CO (November 2003). 11. P. D. Simpson, G. W. Joe, and B. S. Brown, Treatment Retention and Follow-Up Outcomes in the Drug Abuse Treatment Outcomes Study (DATOS), Psychology of Addictive Behaviors 11(4): (1997). 12. The study of therapeutic communities is B. F. Lewis and R. Ross, Retention in Therapeutic Communities: Challenges for the Nineties, in Therapeutic Community: Advances in Research and Application, eds. F. M. Tims, G. DeLeon, and N. Jainchill. NIDA, Rockville, MD (1994). Concerning a baseline for what retention might look like in the worst performing drug courts, in a study of 11 New York State drug courts, the one with the lowest one year retention rate was found to have a rate of 47 percent. Rempel et al., supra note D. Wilson, O. Mitchell, and D.L. MacKenzie, Systematic Review of Drug Court Effects on Recidivism. Paper presented at the Annual Meetings of the American Society of Criminology, Chicago, IL (2002). 14. See S. Aos, P. Phipps, R. Barnoski, and R. Lieb, The Comparative Costs and Benefits of Programs to Reduce Crime Version 4.0. Olympia: Washington State Institute for Public Policy (2001); S. Belenko, Research on Drug Courts: A Critical Review 2001 Update, National Center on Addiction and Substance Abuse, Columbia University (2001); Government Accountability Office, supra note 2; J. Roman and C. DeStefano, Drug Court 17

20 Center for Court Innovation Effects and the Quality of Existing Evidence, in Juvenile Drug Courts and Teen Substance Abuse, eds. J. Butts and J. Roman. Washington, DC: Urban Institute Press (2004). 15. D. C. Gottfredson, B. Kearley, S. S. Najaka, and C. Rocha, The Baltimore City Drug Treatment Court: 3- Year Self-Report Outcome Study, Evaluation Review 29:1 (2005). 16. M. Rempel et al., supra note 9. Results exceeded the statistical margin of error in five of the six sites. The New York study also found significant reductions in recidivism (in five of six sites) over a one-year post-program period, which involves a period of time beginning after the point of drug court graduation and failure. Two other studies have similarly detected post-program reductions in recidivism, A. Bavon, The Effect of the Tarrant County Drug Court Project on Recidivism, 24 Evaluation and Program Planning, 13, 22 (2001); and J. E. Fielding, G. Tye, P. Ogawa, I. J. Imam, A. M. Long, Los Angeles County Drug Court Programs: Initial Results, 23 Journal of Substance Abuse Treatment, 217, 224 (2002). 17. Government Accountability Office, supra note E. Deschenes, S. Turner and P. Greenwood, Drug Court or Probation? An Experimental Evaluation of Maricopa County s Drug Court, 18 The Justice System Journal 55, 73 (1995). 19. D. Gottfredson et al., Effectiveness of Drug Treatment Courts: Evidence from a Randomized Trial, 2 Criminology and Public Policy 171, 196 (2003); A. Harrell, J. Roman, and E. Sack, Drug Court Services for Female Offenders, : Evaluation of the Brooklyn Treatment Court. The Urban Institute, Justice Policy Center. Washington, D.C. (2001). 20. J. Roman and C. DeStefano, supra note 14; and Government Accountability Office, supra note S. Aos et al., supra note S. M. Carey, Evaluating the Cost of California's Drug Courts and Subsequent Policy Effects, Presented at the Annual Conference of the American Society of Criminology, Nashville, TN (November 2004). 23. J. L. Nolan, Reinventing Justice: The American Drug Court Movement, Princeton, N.J.: Princeton University Press (2001). 24. D. Gottfredson et al., supra note M. Rempel et al., supra note M. Rempel and C. D. DeStefano, Predictors of Engagement in Court-Mandated Treatment: Findings at the Brooklyn Treatment Court, , 33 Journal of Offender Rehabilitation, 87, 124 (2001); and M. Rempel et al. supra note M. A. R. Kleiman, Controlling Drug Use and Crime with Testing, Sanctions, and Treatment, in Drug Addiction and Drug Policy: The Struggle to Control Dependence, eds. P. Heymann and W. Brownsberger. Cambridge, MA: Harvard University Press (2001). 28. Anglin et al., supra note 5; DeLeon, supra note 7; Gottfredson et al., supra note 19; R. Peters and M. Murrin, Effectiveness of Treatment-Based Drug Courts in Reducing Criminal Recidivism, Criminal Justice and Behavior 27: (2000); D. D. Simpson et al., Treatment Retention and Follow-Up Outcomes in the Drug Abuse Treatment Outcomes Study (DATOS), 11 Psychology of Addictive Behavior, 294, 307 (1997); K. A. Sirotnik and M. W. Roffe, An Investigation of the Feasibility of Predicting Outcome Indices in the Treatment of Heroin Addiction, 12 International Journal of the Addictions, 755, 775 (1977); F. Taxman, Reducing Recidivism Through A Seamless System of Care: Components of Effective Treatment, Supervision, and Transitional Services in the Community, Greenbelt, MD: Washington/Baltimore HIDTA Project (1998); F. Taxman et al., Treatment as Crime Control: Impact of Substance Abuse Treatment on the Individual Offending Rates of Hard-Core Substance Abusing Offenders, Greenbelt, MD: Washington/Baltimore HIDTA Project (1999). 29. Gottfredson, et al., supra note M. Rempel et al., supra note 9. The differences in use of residential treatment stem partly from variations in the addition severity level of participants, but that factor alone cannot explain the magnitude of the differences. Regarding the tremendous variations in the modalities and quality of services available at treatment programs used by drug courts, see also D. F. Anspach, et al., supra note F. Taxman, Unraveling What Works for Offenders in Substance Abuse Treatment Services, 2 National Drug Court Institute Review, 93, 134 (1999). 32. F. S. Taxman and J. A. Bouffard, Substance Abuse Counselors Treatment Philosophy and the Content of Treatment Services Provided to Offenders in Drug Court Programs, Journal of Substance Abuse Treatment, 25(2), (2003). 18